- Email: [email protected]
Anodic stripping voltammetry of lead with microliter volumes of electrolytes and silver-plated glassy carbon electrodes
Analytica Chinzica Acta. 140 (1982) 59-64 EIsevier Scientific Publishing Company, Amsterdam -
Printed in The Netherlands
ANODIC STRIPPING VOLTAMMETRY OF LEAD WITH MICROLITER VOLUMES OF ELECTROLYTES AND SILVER-PLATED GLASSY CARBON ELECTRODES
T. MIWA, Y. NISHIMURA Faculty
of Engineering,
and A MIZUIKE*
Nagoya
University,
Chikusa-ku.
Nagoya
(Japan)
(Received 9th February 1982)
SUMMARY Nanogram quantities of lead are sufficient precision and rapidity, by carbon microelectrode. This method ppm level in 0.5-l mg of high-purity
determined by anodic stripping voltammetry with using a lOO-~1 electrolyte and a silver-plated glassy is applied to the determination of lead at the low zinc.
Although stripping voltammetry has been extensively studied by many workers, only a few papers have been published to date on stripping voltammetry using electrolytes of ten to several hundred microliters and mercury drop electrodes [I, 21. The present paper describes the use of a silver-plated glassy carbon microelectrode for the anodic stripping voltammetry of nanogram quantities of lead in 100 ~1 of dilute hydrochloric acid electrolytes. Theoretically, the stripping signal with a solid or plated electrode is proportional to m[l - exp(-/St/V)] , where m is the quantity of the analyte, k is a constant, S is the surface area of the working electrode, t is the preelectrolysis time, and V is the electrolyte volume. The stripping background is proportional to the surface area of the working electrode. Therefore, when the other esperimental conditions are f&ed, higher stripping signals and signal-tobackground ratios are obtained with smaller electrolyte volumes. If experimental conditions are optimized, absolute detection limits and rapidity of the determination can be improved without loss of precision, by reducing the electrolyte volume from the conventional milhliter levels to the microliter levels and using an appropriate microelectrode. Another advantage of the miniaturization is economy of high-purity reagents for preparing the electrolytes_ EXPERIMENTAL
Apparatus A Yanaco P8-D polarograph (Yanagimoto Mfg. Co., Kyoto, Japan) was employed for pm-electrolysis and recording of stripping (current-voltage) curves. A clean bench was used to avoid airborne contamination during the preparation of all solutions and the dissolution of samples.
60
The microelectrolysis cell was made of polymethylmethacrylate (Fig. 1; working volume 50-300 ,ul). Its teflon lid was provided with a working electrode and a tapered teflon tubing (1 mm o.d.) for introducing nitrogen. A silver rod (4 mm diameter) or a teflon tube (4 mm o-d.) for a salt bridge was fixed by the thread at the bottom of the cell. Reagents Nitrogen was filtered through a O-l-pm membrane filter. Reagent-grade hydrochloric acid was purified in a vitreous-silica sub-boiling distillation unit. Water was purified by ion exchange, distillation, and adsorption-ion exchange-filtration (Millipore Milli-Q system)_ All other reagents were reagent grade and used without further purification_ Preparation of working electrodes Glassy carbon electrode (GCE). The surface of a glassy carbon rod (Tokai Carbon Mfg. Co., Tokyo, Japan; grade GC-20; 0.9 mm diameter X 20 mm) was polished with a fine emery paper followed by an alumina suspension (0.05pm particle size) to a mirror finish_ After washing with water, one end (ca. 7 mm) of the rod was fixed in a l-mm bore teflon tubing with epoxy resin adhesive at 60-80°C. The electrode was then washed with acetone, 7 M nitric acid and water, successively, with the aid of ultrasonics. Silver-plated glassy carbon electrode [Ag(GC)E]. Silver was plated onto the lower part (length 4.5 mm; surface area 13 mm*) of the GCE at -0.7 V vs. Ag/AgCl in 100 ~1 of 0.3 M hydrochloric acid containing 0.5 pg of silver ion for 3 min with stirring by nitrogen bubbling at a flow rate of 10 ml min-‘. The tip of the teflori tubing for nitrogen was removed from the solution so that the nitrogen stream was used only for purging the cell space. After 30 s, the electrode potential was anodically scanned to 0.0 V vs. Ag/AgCl at a
Teflon
Working dectrode
-Sitver
Fig. 1. Microelectrolysis
rod
cell.
61
rate of 2 V mine*. The above plating and anodic scanning cycle was repeated twice more, except that the last scanning was to +0.8 V vs. Ag/AgCl. During the last scanning, the mat-gray plated surface turned to a metallic luster. The silver on the electrode was determined by atomic absorption spectrometry, and found to be ca. 0.5 pg. The Ag(GC)E was stored in deaerated water. Mercury-plated glassy carbon electrode [Hg(GC)E]. Mercury(H) (10 ~g) was added to 100 ~1 of sample solution (0.25 M KNO,-0.25 M HN03) for in situ mercury film deposition [3]. Instead of a silver rod at the bottom of the microelectrolysis cell, a potassium nitrate salt bridge to an Ag/AgCl reference electrode was used. Silver electrode (AgE). The surface of a silver wire (1 mm diameter X 20 mm, purity 99.99%) was polished with an alumina suspension_ The further treatment was the same as that of the GCE, except that washing with 7 M nitric acid was omitted.
Recommersded
procedure
A loo-,& sample solution (0.6 M hydrochloric acid) was transferred to the microelectroiysis cell. The lower 4.5-mm part of the Ag(GC)E was immersed in the electrolyte, and a pre-electrolysis potential of -0.7 V vs. Ag/AgCl was applied to it for 3 min with stirring by nitrogen bubbling at a flow rate of 10 ml min-‘. The tip of the teflon tubing for nitrogen was then removed from the solution so that the nitrogen stream was used only for purging the cell space. After 30 s, the electrode potential was scanned to 0.0 V vs. Ag/ AgCl at a scan rate of 2 V min-‘, a stripping (current-voltage) curve being recorded_ RESULTS
AND
DISCUSSION
The GCE was initially selected as working electrode, because it has a small background current and a high hydrogen overpotential. However, the stripping peak height of lead was very low, and when the cycle of pre-electrolysis and stripping was repeated, the peak height increased gradually as shown in Fig. 2. This phenomenon was caused by the dissolution of silver from the counter electrode (silver rod at the bottom of the cell) and its deposition on the GCE. Therefore, the Ag(GC)E was tried. As shown in Fig. 3, it gave a high and sharp stripping peak with sufficient reproducibility (relative standard deviation of 5% for 10 ng of lead, n = 15). Similar stripping peaks were obtained with the Ag(GC)E and the Hg(GC)E, both prepared in the sample solution in situ. However, relative standard deviations were larger, i.e. 12% (n = 10) for 10 ng of lead, with these two working electrodes. The AgE gave larger background currents than those obtained with the Ag(GC)E as shown jn Fig. 3. It was suspected that this was due to the absorption of hydro&n atoms in silver during the pre-electrolysis and their oxidation during the stripping_ To confirm this, the background currents of
62
I
0.0
-Q2
-04. VK.
-Q6 -Q8 ns/asa
Fig. 2. Successive increase in peak height of lead during repeated pre-electrolysis and stripping_ Pre-electrolysis on GCE at -0.7 V vs. Ag/AgCl for 3 min in 100 ~1 of 0.1 M I-ICIcontaining 5 ng Pb. The numbers of the run are indicated on the peaks. Fig. 3. Anodic stripping curves for lead. Pre-electrolysis at -0.7 V vs. Ag/AgCl for 3 min in iO0 ~1 of 0.1 M HCI [or 0.25 M KNO,-0.25 M HNO, for Hg(GC)E] containing 5 ng Pb.
the Ag(GC)E and the AgE were measured under various experimental conditions. Figure 4 shows the dependence of the background currents of the Ag(GC)E on silver film thickness and pre-electrolysis time- The silver film (5 ,ug Ag/13 mm*) was saturated with hydrogen atoms within 5 mm of pre-electrolysis. The mechanism was further confirmed by experiments with the AgE as shown in Figs. 5 and 6. The background current caused by sorbed hydrogen in platinum electrodes has been studied by Kolthoff and Tanaka [4] _ From the above results, the Ag(GC)E was found to be the most suitable working electrode for the present work. According to the recommended procedure, about 85% of lead is deposited during the pre-electrolysis. Variations of the electrolyte volume and the electrodeposited area caused by the nitrogen bubbling were practically negligible; the electrolyte surface was sufficiently quiet and the electrolyte volume decreased only Z-4% during the 30-min bubbling. The stripping peak height of lead was independent of the pm-electrolysis potential between -0.6 and -1.0 V vs. Ag/AgCl. The concentration of hydrochloric acid had little effect on the peak height of lead over the range 0.1-0.8 M. A hydrochloric acid concentration of 0.6 M was used in the determinations, because the interference of
63
-0.8
-1.0
v vs. 4/4Cl 22 V mii”
+ Fig. 4. Dependence of background current (pg Ag/13 mm’) and pre-electrolysis time. 0.1 M HCl.
of the Ag(GC)E Pre-electrolysis
Fig. 5. Dependence of background current of electrolysis at -1.0 V vs. Ag/AgCI in 0.1 M HCl.
the
AgE
on
on the at -0.7
Ag film thickness V vs. Ag/AgCl in
pre-electrolysis
time.
Pre-
copper in the determination of lead decreased and the background current increased with increasing acidity. The Ag(GC)E could be used repeatedly about 20 times and stored for over 4 days in oxygen-free water without appreciable change in the peak height. The calibration (peak height vs. concentration) curve was linear from 0.2 ng to at least 20 ng of lead in 100 was
~1
and
about about 0.2 from the antimony,
passed
through
the
origin.
The
relative
standard
deviation
8% at 5 ng of lead, and the lower limit of determination was ng. In the determination of 10 ng of lead, no interference resulted presence of 100 ng of cadmium, and 10 ng each of bismuth, copper and tin.
02Vmir-i’
0.6 VrniR
2 VmiR
Fig. 6. Dependence of background current of the AgE on scan rate. Pre-electrolysis at -0.7 V vs. AglAgCl for 3 min in 0.6 M HCl. (-) Scanning was interrupted for 15-30 s (until steady currents were reached) at the voltages marked with x; (- - - -) scanning was carried out continuously_
64 TABLE
1
Determination
of lead in zinc metal
Sample taken (ms)
Pb found= (ng)
0.49 0.50 0.45b 1.00 1.03 0.98b
4.8 4.8 10.1 11.2 12.5 15.8
“A blank sample.
vaIue (0.5
ng) was subtracted.
Pb in sample @Pm) 9.8 9.6 11.3 11.2 12.3 11.0 Av. 10.9 b5 ng of lead was added
before
dissolution
Determination of lead in zinc metal A 0.5 or 1-mg sample of high-purity zinc (99.99%
of
purity) was dissolved at -50°C in 100 ~1 of 6 M hydrochloric acid in a 2-ml teflon vessel (15 mm 0-d. X 25 mm) with a lid. After complete dissolution of the sample, the solution was evaporated to dryness, the residue was dissolved in 200 ~1 of water, and the solution was evaporated to dryness again. The residue was dissoIved in 100 ~1 of 0.6 M hydrochloric acid, and transferred to the microelectrolysis cell. Lead in the solution was then determined by the recommended procedure. The results are shown in Table 1. The blank value through the entire procedure was about 0.5 ng of lead. The error was about 10% and the time required for a determination was 1.5-2 h. The authors thank Mr. Shue Rubi for his help in the experimental work. REFERENCES 1 2 3 4
W. L. Underkofler and L Shain, Anal. Chem., 33 (1961) 1966. L. Huderovi and K. &ulik. Talanta, 19 (1972) 1285. T. &I_ Florence, J. Electroanal. Chem., 2’7 (1970) 273. I_ hi. Kolthoff and N. Tanaka, Anal. Chem., 26 (1954) 632.
Printed in The Netherlands
ANODIC STRIPPING VOLTAMMETRY OF LEAD WITH MICROLITER VOLUMES OF ELECTROLYTES AND SILVER-PLATED GLASSY CARBON ELECTRODES
T. MIWA, Y. NISHIMURA Faculty
of Engineering,
and A MIZUIKE*
Nagoya
University,
Chikusa-ku.
Nagoya
(Japan)
(Received 9th February 1982)
SUMMARY Nanogram quantities of lead are sufficient precision and rapidity, by carbon microelectrode. This method ppm level in 0.5-l mg of high-purity
determined by anodic stripping voltammetry with using a lOO-~1 electrolyte and a silver-plated glassy is applied to the determination of lead at the low zinc.
Although stripping voltammetry has been extensively studied by many workers, only a few papers have been published to date on stripping voltammetry using electrolytes of ten to several hundred microliters and mercury drop electrodes [I, 21. The present paper describes the use of a silver-plated glassy carbon microelectrode for the anodic stripping voltammetry of nanogram quantities of lead in 100 ~1 of dilute hydrochloric acid electrolytes. Theoretically, the stripping signal with a solid or plated electrode is proportional to m[l - exp(-/St/V)] , where m is the quantity of the analyte, k is a constant, S is the surface area of the working electrode, t is the preelectrolysis time, and V is the electrolyte volume. The stripping background is proportional to the surface area of the working electrode. Therefore, when the other esperimental conditions are f&ed, higher stripping signals and signal-tobackground ratios are obtained with smaller electrolyte volumes. If experimental conditions are optimized, absolute detection limits and rapidity of the determination can be improved without loss of precision, by reducing the electrolyte volume from the conventional milhliter levels to the microliter levels and using an appropriate microelectrode. Another advantage of the miniaturization is economy of high-purity reagents for preparing the electrolytes_ EXPERIMENTAL
Apparatus A Yanaco P8-D polarograph (Yanagimoto Mfg. Co., Kyoto, Japan) was employed for pm-electrolysis and recording of stripping (current-voltage) curves. A clean bench was used to avoid airborne contamination during the preparation of all solutions and the dissolution of samples.
60
The microelectrolysis cell was made of polymethylmethacrylate (Fig. 1; working volume 50-300 ,ul). Its teflon lid was provided with a working electrode and a tapered teflon tubing (1 mm o.d.) for introducing nitrogen. A silver rod (4 mm diameter) or a teflon tube (4 mm o-d.) for a salt bridge was fixed by the thread at the bottom of the cell. Reagents Nitrogen was filtered through a O-l-pm membrane filter. Reagent-grade hydrochloric acid was purified in a vitreous-silica sub-boiling distillation unit. Water was purified by ion exchange, distillation, and adsorption-ion exchange-filtration (Millipore Milli-Q system)_ All other reagents were reagent grade and used without further purification_ Preparation of working electrodes Glassy carbon electrode (GCE). The surface of a glassy carbon rod (Tokai Carbon Mfg. Co., Tokyo, Japan; grade GC-20; 0.9 mm diameter X 20 mm) was polished with a fine emery paper followed by an alumina suspension (0.05pm particle size) to a mirror finish_ After washing with water, one end (ca. 7 mm) of the rod was fixed in a l-mm bore teflon tubing with epoxy resin adhesive at 60-80°C. The electrode was then washed with acetone, 7 M nitric acid and water, successively, with the aid of ultrasonics. Silver-plated glassy carbon electrode [Ag(GC)E]. Silver was plated onto the lower part (length 4.5 mm; surface area 13 mm*) of the GCE at -0.7 V vs. Ag/AgCl in 100 ~1 of 0.3 M hydrochloric acid containing 0.5 pg of silver ion for 3 min with stirring by nitrogen bubbling at a flow rate of 10 ml min-‘. The tip of the teflori tubing for nitrogen was removed from the solution so that the nitrogen stream was used only for purging the cell space. After 30 s, the electrode potential was anodically scanned to 0.0 V vs. Ag/AgCl at a
Teflon
Working dectrode
-Sitver
Fig. 1. Microelectrolysis
rod
cell.
61
rate of 2 V mine*. The above plating and anodic scanning cycle was repeated twice more, except that the last scanning was to +0.8 V vs. Ag/AgCl. During the last scanning, the mat-gray plated surface turned to a metallic luster. The silver on the electrode was determined by atomic absorption spectrometry, and found to be ca. 0.5 pg. The Ag(GC)E was stored in deaerated water. Mercury-plated glassy carbon electrode [Hg(GC)E]. Mercury(H) (10 ~g) was added to 100 ~1 of sample solution (0.25 M KNO,-0.25 M HN03) for in situ mercury film deposition [3]. Instead of a silver rod at the bottom of the microelectrolysis cell, a potassium nitrate salt bridge to an Ag/AgCl reference electrode was used. Silver electrode (AgE). The surface of a silver wire (1 mm diameter X 20 mm, purity 99.99%) was polished with an alumina suspension_ The further treatment was the same as that of the GCE, except that washing with 7 M nitric acid was omitted.
Recommersded
procedure
A loo-,& sample solution (0.6 M hydrochloric acid) was transferred to the microelectroiysis cell. The lower 4.5-mm part of the Ag(GC)E was immersed in the electrolyte, and a pre-electrolysis potential of -0.7 V vs. Ag/AgCl was applied to it for 3 min with stirring by nitrogen bubbling at a flow rate of 10 ml min-‘. The tip of the teflon tubing for nitrogen was then removed from the solution so that the nitrogen stream was used only for purging the cell space. After 30 s, the electrode potential was scanned to 0.0 V vs. Ag/ AgCl at a scan rate of 2 V min-‘, a stripping (current-voltage) curve being recorded_ RESULTS
AND
DISCUSSION
The GCE was initially selected as working electrode, because it has a small background current and a high hydrogen overpotential. However, the stripping peak height of lead was very low, and when the cycle of pre-electrolysis and stripping was repeated, the peak height increased gradually as shown in Fig. 2. This phenomenon was caused by the dissolution of silver from the counter electrode (silver rod at the bottom of the cell) and its deposition on the GCE. Therefore, the Ag(GC)E was tried. As shown in Fig. 3, it gave a high and sharp stripping peak with sufficient reproducibility (relative standard deviation of 5% for 10 ng of lead, n = 15). Similar stripping peaks were obtained with the Ag(GC)E and the Hg(GC)E, both prepared in the sample solution in situ. However, relative standard deviations were larger, i.e. 12% (n = 10) for 10 ng of lead, with these two working electrodes. The AgE gave larger background currents than those obtained with the Ag(GC)E as shown jn Fig. 3. It was suspected that this was due to the absorption of hydro&n atoms in silver during the pre-electrolysis and their oxidation during the stripping_ To confirm this, the background currents of
62
I
0.0
-Q2
-04. VK.
-Q6 -Q8 ns/asa
Fig. 2. Successive increase in peak height of lead during repeated pre-electrolysis and stripping_ Pre-electrolysis on GCE at -0.7 V vs. Ag/AgCl for 3 min in 100 ~1 of 0.1 M I-ICIcontaining 5 ng Pb. The numbers of the run are indicated on the peaks. Fig. 3. Anodic stripping curves for lead. Pre-electrolysis at -0.7 V vs. Ag/AgCl for 3 min in iO0 ~1 of 0.1 M HCI [or 0.25 M KNO,-0.25 M HNO, for Hg(GC)E] containing 5 ng Pb.
the Ag(GC)E and the AgE were measured under various experimental conditions. Figure 4 shows the dependence of the background currents of the Ag(GC)E on silver film thickness and pre-electrolysis time- The silver film (5 ,ug Ag/13 mm*) was saturated with hydrogen atoms within 5 mm of pre-electrolysis. The mechanism was further confirmed by experiments with the AgE as shown in Figs. 5 and 6. The background current caused by sorbed hydrogen in platinum electrodes has been studied by Kolthoff and Tanaka [4] _ From the above results, the Ag(GC)E was found to be the most suitable working electrode for the present work. According to the recommended procedure, about 85% of lead is deposited during the pre-electrolysis. Variations of the electrolyte volume and the electrodeposited area caused by the nitrogen bubbling were practically negligible; the electrolyte surface was sufficiently quiet and the electrolyte volume decreased only Z-4% during the 30-min bubbling. The stripping peak height of lead was independent of the pm-electrolysis potential between -0.6 and -1.0 V vs. Ag/AgCl. The concentration of hydrochloric acid had little effect on the peak height of lead over the range 0.1-0.8 M. A hydrochloric acid concentration of 0.6 M was used in the determinations, because the interference of
63
-0.8
-1.0
v vs. 4/4Cl 22 V mii”
+ Fig. 4. Dependence of background current (pg Ag/13 mm’) and pre-electrolysis time. 0.1 M HCl.
of the Ag(GC)E Pre-electrolysis
Fig. 5. Dependence of background current of electrolysis at -1.0 V vs. Ag/AgCI in 0.1 M HCl.
the
AgE
on
on the at -0.7
Ag film thickness V vs. Ag/AgCl in
pre-electrolysis
time.
Pre-
copper in the determination of lead decreased and the background current increased with increasing acidity. The Ag(GC)E could be used repeatedly about 20 times and stored for over 4 days in oxygen-free water without appreciable change in the peak height. The calibration (peak height vs. concentration) curve was linear from 0.2 ng to at least 20 ng of lead in 100 was
~1
and
about about 0.2 from the antimony,
passed
through
the
origin.
The
relative
standard
deviation
8% at 5 ng of lead, and the lower limit of determination was ng. In the determination of 10 ng of lead, no interference resulted presence of 100 ng of cadmium, and 10 ng each of bismuth, copper and tin.
02Vmir-i’
0.6 VrniR
2 VmiR
Fig. 6. Dependence of background current of the AgE on scan rate. Pre-electrolysis at -0.7 V vs. AglAgCl for 3 min in 0.6 M HCl. (-) Scanning was interrupted for 15-30 s (until steady currents were reached) at the voltages marked with x; (- - - -) scanning was carried out continuously_
64 TABLE
1
Determination
of lead in zinc metal
Sample taken (ms)
Pb found= (ng)
0.49 0.50 0.45b 1.00 1.03 0.98b
4.8 4.8 10.1 11.2 12.5 15.8
“A blank sample.
vaIue (0.5
ng) was subtracted.
Pb in sample @Pm) 9.8 9.6 11.3 11.2 12.3 11.0 Av. 10.9 b5 ng of lead was added
before
dissolution
Determination of lead in zinc metal A 0.5 or 1-mg sample of high-purity zinc (99.99%
of
purity) was dissolved at -50°C in 100 ~1 of 6 M hydrochloric acid in a 2-ml teflon vessel (15 mm 0-d. X 25 mm) with a lid. After complete dissolution of the sample, the solution was evaporated to dryness, the residue was dissolved in 200 ~1 of water, and the solution was evaporated to dryness again. The residue was dissoIved in 100 ~1 of 0.6 M hydrochloric acid, and transferred to the microelectrolysis cell. Lead in the solution was then determined by the recommended procedure. The results are shown in Table 1. The blank value through the entire procedure was about 0.5 ng of lead. The error was about 10% and the time required for a determination was 1.5-2 h. The authors thank Mr. Shue Rubi for his help in the experimental work. REFERENCES 1 2 3 4
W. L. Underkofler and L Shain, Anal. Chem., 33 (1961) 1966. L. Huderovi and K. &ulik. Talanta, 19 (1972) 1285. T. &I_ Florence, J. Electroanal. Chem., 2’7 (1970) 273. I_ hi. Kolthoff and N. Tanaka, Anal. Chem., 26 (1954) 632.